sysbench onednn icelake

Intel Core i7-1065G7 testing with a Dell 06CDVY (1.0.9 BIOS) and Intel Iris Plus G7 3GB on Ubuntu 20.10 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2103134-HA-SYSBENCHO81&grs&sor.

sysbench onednn icelakeProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen Resolution123Intel Core i7-1065G7 @ 3.90GHz (4 Cores / 8 Threads)Dell 06CDVY (1.0.9 BIOS)Intel Ice Lake-LP DRAM16GBToshiba KBG40ZPZ512G NVMe 512GBIntel Iris Plus G7 3GB (1100MHz)Realtek ALC289Intel Killer Wi-Fi 6 AX1650i 160MHzUbuntu 20.105.9.0-050900-generic (x86_64)GNOME Shell 3.38.2X Server 1.20.94.6 Mesa 20.2.61.2.145GCC 10.2.0ext41920x1200OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: intel_pstate powersave - CPU Microcode: 0xa0 - Thermald 2.3 Security Details- itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

sysbench onednn icelakeonednn: IP Shapes 3D - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUsysbench: CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUsysbench: RAM / Memoryonednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: IP Shapes 3D - bf16bf16bf16 - CPUonednn: IP Shapes 1D - bf16bf16bf16 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUonednn: IP Shapes 1D - f32 - CPU1239.314785.830563.3611816.618583.703968.461312670.46565.114.697119839.216633.2312895.522.30136709.1212.570213325.615571.5454.32834.0408914.822518.106712.271936.95814.415953.7842315.647368.044466.388423.7290017.753489.809272.602113341.86943.264.447039311.076939.8813254.322.05936842.2612.733213255.014396.1059.08013.8594314.994017.180311.5432134.43324.147083.453769.395678.031466.504843.7243817.940090.286973.386713445.06954.154.705589326.286934.9313425.822.87866944.8412.971713492.614220.1659.22824.1830815.774716.472211.3227734.41224.164863.249229.62702OpenBenchmarking.org

oneDNN

Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU3213691215SE +/- 0.06519, N = 3SE +/- 0.07005, N = 3SE +/- 0.08928, N = 158.031468.044469.31478MIN: 6.52MIN: 6.49MIN: 6.61. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU123246810SE +/- 0.01898, N = 3SE +/- 0.08530, N = 3SE +/- 0.01171, N = 35.830566.388426.50484MIN: 4.06MIN: 4.36MIN: 4.551. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU1320.8391.6782.5173.3564.195SE +/- 0.03652, N = 3SE +/- 0.00171, N = 3SE +/- 0.00207, N = 33.361183.724383.72900MIN: 2.28MIN: 2.53MIN: 2.511. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU12348121620SE +/- 0.11, N = 3SE +/- 0.21, N = 3SE +/- 0.01, N = 316.6217.7517.94MIN: 14.21MIN: 15.37MIN: 15.571. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU12320406080100SE +/- 0.46, N = 3SE +/- 0.58, N = 15SE +/- 0.69, N = 383.7089.8190.29MIN: 75.71MIN: 75.56MIN: 79.61. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU1231632486480SE +/- 1.07, N = 13SE +/- 0.96, N = 12SE +/- 0.95, N = 1268.4672.6073.39MIN: 43.78MIN: 44.41MIN: 44.451. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU1233K6K9K12K15KSE +/- 55.86, N = 3SE +/- 98.80, N = 3SE +/- 24.56, N = 312670.413341.813445.0MIN: 12140.7MIN: 12941.9MIN: 13241.11. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU12315003000450060007500SE +/- 47.71, N = 3SE +/- 7.44, N = 3SE +/- 13.67, N = 36565.116943.266954.15MIN: 6214.93MIN: 6825.81MIN: 6811.341. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU2131.05882.11763.17644.23525.294SE +/- 0.02905, N = 3SE +/- 0.03979, N = 15SE +/- 0.05221, N = 54.447034.697114.70558MIN: 3.56MIN: 3.58MIN: 3.661. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Sysbench

Test: CPU

OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPU1322K4K6K8K10KSE +/- 78.19, N = 3SE +/- 78.44, N = 3SE +/- 43.55, N = 39839.219326.289311.071. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU13215003000450060007500SE +/- 18.80, N = 3SE +/- 5.85, N = 3SE +/- 19.43, N = 36633.236934.936939.88MIN: 6313.9MIN: 6814.8MIN: 6790.851. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU1233K6K9K12K15KSE +/- 125.57, N = 3SE +/- 119.03, N = 3SE +/- 54.42, N = 312895.513254.313425.8MIN: 12104.2MIN: 12527.8MIN: 13175.21. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU213510152025SE +/- 0.21, N = 15SE +/- 0.21, N = 3SE +/- 0.24, N = 1522.0622.3022.88MIN: 15.55MIN: 15.43MIN: 15.521. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU12315003000450060007500SE +/- 69.49, N = 3SE +/- 81.19, N = 3SE +/- 6.59, N = 36709.126842.266944.84MIN: 6404.62MIN: 6444.48MIN: 6816.791. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU1233691215SE +/- 0.03, N = 3SE +/- 0.13, N = 6SE +/- 0.11, N = 312.5712.7312.97MIN: 11.29MIN: 11.27MIN: 11.281. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU2133K6K9K12K15KSE +/- 109.13, N = 3SE +/- 52.75, N = 3SE +/- 28.89, N = 313255.013325.613492.6MIN: 12425.1MIN: 12689.5MIN: 13267.61. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Sysbench

Test: RAM / Memory

OpenBenchmarking.orgMiB/sec, More Is BetterSysbench 1.0.20Test: RAM / Memory1233K6K9K12K15KSE +/- 176.07, N = 15SE +/- 213.47, N = 15SE +/- 285.42, N = 1415571.5414396.1014220.161. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU1231326395265SE +/- 0.68, N = 15SE +/- 0.94, N = 15SE +/- 0.94, N = 1454.3359.0859.23MIN: 45.3MIN: 45.45MIN: 45.961. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU2130.94121.88242.82363.76484.706SE +/- 0.06554, N = 15SE +/- 0.08742, N = 12SE +/- 0.08071, N = 133.859434.040894.18308MIN: 2.76MIN: 2.76MIN: 2.791. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU12348121620SE +/- 0.21, N = 15SE +/- 0.24, N = 15SE +/- 0.28, N = 1514.8214.9915.77MIN: 11.78MIN: 11.78MIN: 11.881. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU32148121620SE +/- 0.05, N = 3SE +/- 0.26, N = 15SE +/- 0.34, N = 1316.4717.1818.11MIN: 12.36MIN: 12.28MIN: 12.271. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU3213691215SE +/- 0.13, N = 15SE +/- 0.19, N = 15SE +/- 0.17, N = 311.3211.5412.27MIN: 6.15MIN: 5.86MIN: 6.021. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU321816243240SE +/- 0.68, N = 12SE +/- 0.69, N = 12SE +/- 0.44, N = 1234.4134.4336.96MIN: 22.57MIN: 22.52MIN: 22.661. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU2310.99361.98722.98083.97444.968SE +/- 0.06809, N = 15SE +/- 0.08889, N = 12SE +/- 0.07740, N = 154.147084.164864.41595MIN: 2.41MIN: 2.43MIN: 2.41. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU3210.85151.7032.55453.4064.2575SE +/- 0.17512, N = 12SE +/- 0.11530, N = 12SE +/- 0.04166, N = 153.249223.453763.78423MIN: 1.75MIN: 1.77MIN: 21. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

oneDNN

Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU23148121620SE +/- 0.08234, N = 3SE +/- 0.11395, N = 3SE +/- 0.47191, N = 149.395679.6270215.64736MIN: 7.87MIN: 8.11MIN: 8.61. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl


Phoronix Test Suite v10.8.5